R e s e a rc h

A review on biomass torrefaction process

and product properties for energy
applications
Jaya Shankar Tumuluru,1* Shahab Sokhansanj,2 J. Richard Hess,1 Christopher T. Wright,1 and Richard D. Boardman1

Abstract
Torrefaction of biomass can be described as a mild form of pyrolysis at temperatures typically ranging between 200 and 300°C in an inert and
reduced environment. Common biomass reactions during torrefaction include devolatilization, depolymerization, and carbonization of hemicellulose,
lignin, and cellulose. The torrefaction process produces a brown to black uniform solid product, as well as condensable (water, organics, and lipids) and
­noncondensable gases (CO2, CO, and CH4). Typically during torrefaction, 70% of the mass is retained as a solid product, containing 90% of the initial
energy content, while 30% of the lost mass is converted into condensable and noncondensable products. The system’s energy efficiency can be improved
by reintroducing the material lost during torrefaction as a source of heat. Torrefaction of biomass improves its physical properties like grindability; par-
ticle shape, size, and distribution; pelletability; and proximate and ultimate composition like moisture, carbon and hydrogen content, and calorific value.
Compared to raw biomass, the carbon content and calorific value of torrefied biomass increases by 15–25% wt, while the moisture content decreases to
<3% (w.b.). Torrefaction decreases the grinding energy by about 70%, and the ground torrefied biomass has improved sphericity, particle surface area, and
particle size distribution. Torrefied biomass pelletization at temperatures of 225°C decreases the specific energy consumption and increases the capacity
of the mill by a factor of 2. The loss of the OH functional group during torrefaction makes the material hydrophobic (i.e., loses the ability to attract water
molecules) and more stable against chemical oxidation and microbial degradation. These improved properties make torrefied biomass particularly suitable
for cofiring in power plants and as an upgraded feedstock for gasification.

Idaho National Laboratory, 2525 Fremont Avenue, Idaho Falls, Idaho 83415,
1
USA 2Oak Ridge National Laboratory, PO Box 2008, Oak Ridge, Tennesse
37831, USA
*Author for correspondence. Email: jayashankar.tumuluru@inl.gov
Submitted: 5 August 2011; Revised: 16 September 2011; Accepted: 17
September 2011
DOI: 10.1089/ind.2011.0014
Keywords: biomass torrefaction; torrefaction temperature and residence
time; biomass reactions; off-gas composition; torrefied biomass physical,
chemical, and storage properties
ABBREVIATIONS: CV, calorific value; GHG, greenhouse gas; TOP, torrefaction
and pelletization
Introduction
from different thermochemical (combustion, gasification, and pyrolysis),
biological (anaerobic digestion and ­fermentation), or chemical (esteri-
fication) processes, where direct combustion can provide a near-term
energy solution.1 The growing interest in biomass as a solid fuel includes
combustion to produce steam for electrical power and commercial plant
uses, as well as gasification to produce a combustible gas (large partial
pressure of nitrogen and CO2, called producer gas) and syngas (carbon
monoxide and hydrogen with low amounts of nitrogen and CO2). Still,
the use of either producer gas or syngas in modern reciprocating or gas
turbines, or to produce higher value chemicals and fuels, is limited due to
biomass feedstock preparation, accumulation logistics, and economics.
Limitations of biomass as fuel
Some of the inherent problems with raw biomass materials

T
he world is currently challenged to reduce dependence on fos-
sil fuels and achieve a sustainable, renewable ­energy supply.
Energy produced from biomass is considered ­carbon-neutral
because the carbon dioxide released during conversion is
already part of the carbon cycle.1 Increasing biomass use for energy can
help reduce greenhouse gas (GHG) emissions and meet the targets estab-
lished in the Kyoto Protocol.2 Energy from biomass can be produced
c­ ompared to fossil fuel resources (low bulk density, high moisture
content, hydrophilic nature, and low calorific value [CV])  render
raw biomass difficult to use on a large scale.1 These limitations
greatly impact logistics and final energy efficiency. Due to its
low energy density compared to fossil fuels, very high volumes of
biomass are needed, which compounds problems associated with
storage, transportation, and feed handling at cogeneration, thermo-
chemical, and biochemical conversion plants.

384 INdustrial Biotechnology october 2011


High moisture in raw biomass is one of the primary challenges,
as it reduces the efficiency of the process and increases fuel pro-
duction costs.3 High moisture content in biomass leads to natural
decomposition, resulting in loss of quality and storage issues such
as off-gas emissions. Another consequence of high moisture ­content
is the uncertainty it causes in biomass’s physical, chemical, and
microbiological properties. Irregular biomass shapes constitute
another issue, especially during feeding in a cofiring or gasification
system. In addition, biomass has more oxygen than carbon and
hydrogen, making it less suitable for thermochemical conversion
processes. Considered collectively, these properties make raw bio-
mass ­unacceptable for energy applications.
To overcome these challenges and make biomass suitable for
energy applications, the material must be preprocessed. One of the
commonly used preprocessing operations is grinding, which helps to
achieve a consistent particle size; however, the moisture content of
the biomass limits the performance of many grinders.3 Furthermore,
grinding can be very costly when smaller particle sizes are desired
and, in some cases, impractical for biomass with high moisture con-
tent. High moisture content can also result in inconsistent particle
sizes (especially when the particles are less than 2 mm), which may
not react consistently, thereby reducing the efficiency and increasing
the costs of the conversion process. Also, raw biomass is thermally
unstable due to high moisture, which results in low CVs and incon-
sistent particle-size distribution issues when used in thermochemical
processes such as gasification. This can lead to inconsistent products
and the formation of condensable tars, which results in problems like
gas-line blockage.4
A viable option is to pretreat the biomass before the end-use
application. Pretreatment helps alter biomass’s physical properties
and chemical composition and makes it more suitable for conver-
sion.5 The pretreatment can be a chemical, thermal, or mechanical
process, like ammonia fiber explosion, torrefaction, and steam
explosion, respectively. These pretreatment processes help alter the
amorphous and crystalline regions of the biomass and bring sig-
nificant changes in structural and chemical compositions. Figure 1
Cellulose
Lignin
Amorphous Pretreatment
region
Crystalline
region
Hemicellulose
Figure 1. Pretreatment effect on lignocellulosic biomass 5
© mary ann liebert, inc.
research
shows how the pretreatment of biomass makes the biomass easier
to convert.5
Torrefaction, which is a thermal pretreatment process, is a viable
technology that significantly alters the physical and chemical com-
position of the biomass. Torrefaction is defined as slowly heating bio-
mass in an inert environment and temperature range of ­200–300°C.
This  process improves the physical, chemical, and biochemical
composition of the biomass, making it perform better for cofiring
and gasification purposes. Many researchers have studied the effect
of torrefaction process time and temperature on the physical and
chemical composition.6–15 However, a detailed literature review is
lacking on the torrefaction process in terms of biomass reactions
(such as depolymerization, devolatilization, and carbonization) and
product properties. The primary focus of this research is to conduct
a detailed literature review on biomass ­torrefaction, which includes
(a) biomass reactions, including chemical and structural changes, (b)
torrefaction product yields in terms of condensable, noncondens-
able, and solid product, and (c) the ­solid torrefied product’s physical,
chemical, and storage properties for energy applications.
Biomass components
The plant cell wall is the tough, usually flexible but sometimes
fairly rigid layer that provides structural support and protection
from mechanical and thermal stresses. The major components of
the primary cell wall are cellulose (carbohydrates), hemicelluloses,
and pectin. The cellulose microfibrils are linked via hemicellu-
losic ­tethers to form the cellulose-hemicellulose network, which is
embedded in the pectin matrix. The outer part of the primary cell
wall is usually impregnated with cutin and wax, forming a perme-
ability barrier known as the plant cuticle.16
Cells and tissues of the plant body play an important role in the
growth of the plant. The structural complexity of the plant body
results from variations in the form and function of the cells and also
from differences in the manner of combination of cells into tissue
and a tissue system. The three different types of plant tissues are
(1) meristematic tissue, (2) dermal tissue, and (3) vascular tissue.17
Secondary cell walls contain a wide range of additional ­compounds
that modify their mechanical properties and ­p ermeability. The
­polymers that make up the secondary cell wall include (1) ­cellulose,
(2) xylan, a type of hemicellulose, (3) lignin, a complex phenolic
polymer that penetrates the spaces in the cell wall between cel-
lulose, hemicellulose, and pectin components and which drives out
water and strengthens the wall, and (4) structural proteins (approxi-
mately 1–5%), which are found in most plant cell walls.16 Figure 2
shows the plant cell wall and lignocellulosic biomass composition.5
Table 1 shows the typical lignocellulosic content of some plant and
woody biomass.12,18
Torrefaction process overview
Torrefaction is a thermal pretreatment technology. It is also defined
as isothermal pyrolysis of biomass occurring in temperature ranges of
200–300°C and performed at atmospheric pressure in the absence of
• Vol. 7 No. 5 • october 2011 INdustrial Biotechnology 385
research

Plant cell wall

Cellulose

Lignin

Hemicellulose

Figure 2. Plant cell wall and lignocellulosic biomass composition5

reaction is more exothermal, and the degradation of hemicellulose

Table 1. Typical lignocellulosic content of some plant
continues. At these temperatures, biomass begins to brown and give
materials12,18
off additional moisture, CO2, and large amounts of acetic acid with
Lignocellulosic content (%) some phenols that have low energy values.7
Plant material Hemicellulose Cellulose Lignin During torrefaction, the major decomposition reactions affect the
hemicellulose. Lignin and cellulose also decompose in the range of
Orchard grass
40.0 32.0 4.7 temperatures at which torrefaction is normally carried out, but to a
(medium maturity)
lesser degree.19–20 Torrefied biomass retains most of its energy and
Rice straw 27.2 34.0 14.2 simultaneously loses its hygroscopic properties. At about 280°C,
Birch wood 25.7 40.0 15.7 the reaction is entirely exothermic, and gas production increases,
resulting in the formation of carbon monoxide, hydrocarbons like
Reed canary grass 29.7 42.6 7.6
phenols and cresols, and other heavier products. For torrefaction,
Wheat straw 30.8 41.3 7.7 process temperatures over 300°C are not recommended, as these
Willow 14.1 49.3 20.0 may lead to extensive devolatilization of the biomass due to the
initiation of the pyrolysis process.8

Torrefaction process technique

oxygen.6–7 Biomass torrefaction has been recognized as a technically
feasible method of converting raw biomass into a solid that is suitable
for commercial and residential combustion and gasification applica-
tions, given that it has high energy density, is hydrophobic, com-
pactable, and grindable, and has a lower oxygen-to-carbon (O/C) ratio.
During the initial heating of lignocellulosic materials, water due
to chemical reactions is removed through a thermocondensation
process. This happens at temperatures between 160 and 180°C and
results in the formation of CO2.7 At temperatures of 180–270°C, the
Torrefaction is a thermochemical process involving the interac-
tion of drying and incomplete pyrolysis. The different parameters
that influence the torrefaction process are (a) reaction temperature,
(b) heating rate, (c) absence of oxygen, (d) residence time, (e) ­ambient
pressure, (d) flexible feedstock, (e) feedstock moisture, and (f) feed-
stock particle size. Biomass feedstock is typically predried to 10%
or less moisture content prior to torrefaction. Particle size plays
an important role in torrefaction in that it influences the reaction
mechanisms, kinetics, and duration of the process, given a specific

386 INdustrial Biotechnology october 2011

 research

heating rate. The chemical reactions that occur when reactive inter-
mediates are trapped in a thick matrix differ from the situations
in which products can escape and be swept away in a gas stream.
The duration of the process is basically adjusted to produce friable,
hydrophobic, and energy-rich enhanced biomass fuel.
Temperature effect on the biomass components
Understanding plant components and their composition will
help in understanding the biomass degradation reactions during
the thermal pretreatment process. Exposure of biomass to elevated
temperatures results in thermal degradation of its structure, which
is often accompanied by loss of mass. The degree of thermal degra-
dation depends on the duration of the heating and temperature.21–23
The various components in the biomass perform unique functions
within the plant structure and interact differently depending on
thermal treatment time and temperature.
Drying process variables that can influence the structural and
chemical compositional changes include biomass composition,
particle size, processing temperature and time, heating rate, gas
composition, pressure, and flow rate.9 Figure 3 provides a modified
version of Bergman et al’s description of the torrefaction process.10
The updated figure describes the structural changes, emissions due
to bond cleavage, biomass reactions, and color changes at different
temperature regimes ranging from 50–300°C.9,21–30

Reactive drying (initiates

Nonreactive drying (no changes changes in chemical Destructive drying (alters
in chemical composition) composition) chemical composition)
Mostly surface Insignificant organic Initiation of hydrogen Breakage of inter- and intramolecular hydrogen,
Water, organic emissions,

moisture removal emissions and carbon bonds C-O and C-C bonds. Emission of hydrophilic
breaking. Emission extractives (organic liquid product having
of lipophylic compounds oxygenated compounds). Formation of higher
and gases

like saturated and molecular mass carboxylic acids

unsaturated fatty acids, (CH3-(CH2)n-COOH), n=10-30), alcohols,
sterols, terpenes, which aldehydes, ether and gases like CO, CO2 and CH4
have no capacity to form
hydrogen bonds.

Initial disruption of Maximum cell Structural Complete destruction of cell structure. Biomass
Cell and

cell structure structure disruption deformity loses its fibrous nature and acts very brittle.
tissue

and reduced
porosity
Hemicellulose

Extensive
Depolymerization Limited
devolatilization
Drying (A) and devolatilization and
recondensation and carbonization
(C) carbonization (D) (E)
Lignin

Glass transition/
A C D E
softening (B)
Cellulose

A C D E
Color changes
in biomass

Torrefaction

50 100 150 200 250 300

Temperature (°C)
Figure 3. Structural, chemical, and color changes in biomass at different drying temperatures (modified version of Bergman et al’s ­description
of the torrefaction process10)

© mary ann liebert, inc. • Vol. 7 No. 5 • october 2011 INdustrial Biotechnology 387
research
Based on Figure 3, at drying temperatures of 50–150°C (A), ­biomass
loses moisture and shrinks. This also results in reduced porosity in the
biomass, though the biomass may still have the ability to retain its
structure if rewetted. This region is known as the nonreactive drying
zone, where most of the chemical constituents of the biomass remain
intact. At the higher end of these temperatures (i.e., 120–150°C)
(B), the lignin softens and makes the material more suitable for
densification, as the softened lignin acts as a binder. Temperature
Regime C (i.e., 150–200°C), also called the reactive drying range,
initiates the breakage of hydrogen and carbon bonds and results in
the emission of lipophilic extractives and compounds due to thermal
degradation of biomass solids. This temperature also results in struc-
tural ­deformity, from which biomass loses its ability to regain its
original structure if ­rewetted. Also, according to Bergman and Kiel,
­depolymerization of ­hemicellulose results in shortened, condensed
polymers with solid structures.10,31 Increasing the temperature further,
as shown in Regime D, also called destructive drying (200–300°C),
results in carbonization and devolatilization.
These temperatures represent the torrefaction process limits,
which result in the disruption of most inter- and intramolecular
hydrogen bonds and C–C and C–O bonds, resulting in the formation
of hydrophilic extractives, carboxylic acids, alcohols, aldehydes,
ether, and gases like CO, CO2, and CH4. At these temperatures, cell
structure is completely destroyed as the biomass loses its fibrous
nature and becomes brittle. Bergman reports further that increas-
ing the temperature to >300°C results in extensive devolatilization
and carbonization of the polymers.10,32 The blue line in Figure 3
indicates that at temperatures lower than 250°C, the mass loss is at
a ­minimum, as main biomass decomposition results from limited
devolatilization and carbonization of the hemicellulose. At tem-
peratures >250°C, the hemicellulose decomposes extensively into
volatiles and a char-like solid product. Lignin and cellulose show
limited devolatilization and carbonization.
Color change in the raw material can be a good means of describ-
ing the degree of torrrefaction. The color changes in the biomass at
Table 2. Thermal treatment process variables for different temperature regimes
TEMPERATURE Time Process reactions
(°C) (min)
50–150 30–120 Nonreactive drying (moisture removal and structural
changes)
150–200 30–120 Reactive drying (moisture removal and structural
damage due to cell wall collapse)
200–300 <30 Destructive drying
Ø Devolatilization and carbonization of hemicellulose
Ø Depolymerization and devolatilization/softening
of lignin
Ø Depolymerization and devolatilization of cellulose
388 INdustrial Biotechnology october 2011
different temperature regimes indicate that the biomass turns brown
to black at 150–300°C, which can be mainly attributed to chemical
compositional changes. Lam et al quantified the severity of steam
treatment based on color coordinates (such as L, a, and b values,
which indicate lightness, redness/greenness, and yellowness/blue-
ness, respectively) in their study.33 The same authors also developed
multilinear regression models to describe the chemical composi-
tional changes like carbon and hydrogen based on color changes in
steam-exploded wood pellets.33 In addition, color measurement can
be a good indicator for identifying impurities like bark, ash, or any
other foreign material present in the biomass. For example, in the
coffee bean roasting process, the change in the color is used as an
indicator to define the changes in chemical composition.34 Typical
thermal treatment process variables (mass and energy yields at dif-
ferent temperature regimes) are given in Table 2.
During torrefaction at 200–300°C, mass loss predominantly
results from the loss of moisture and decomposition (devolatiliza-
tion), particularly hemicellulose and some lignin. Xylan-based
hemicellulose generally decomposes around 250–280°C. Lignin
decomposition proceeds more slowly, but gradually increases start-
ing at about 200°C.11 However, the thermal decomposition behavior
of individual biomass polymers may be different from the strongly
integrated structure of whole biomass. Figure 4 indicates the typical
weight loss in cottonwood as a function of temperature.35 The figure
clearly shows that the weight loss and degradation of biochemical
composition typically starts at temperatures higher than 200°C.
Biomass reactions
Cell and tissue: Changes in biomass cells and tissues, which lead
to structural changes, typically happen at drying temperatures of
50–150°C. Terziev observed that different thermal treatments have a
distinct effect on wood microstructure and properties.36 Some studies
on drying wood in a 100–150°C temperature range indicated signifi-
cant cell wall shrinkage and a pore size volume decrease, which is
attributed to moisture loss.37 The drying damage manifests as irregu-
Heating rate Drying Mass yield Energy yield
(°C/min) environment (%) (%)
and pressure
<50 Air and ambient ~90–95 Not
pressure significant
<50 Air and ambient ~90 Needs to
pressure be researched
<50 Inert ~70 ~90
environment
and ambient
pressure

larly distributed micro-cracks within the biomass cell wall. Three
mechanisms result in structural changes to the biomass due to drying.
First, initially elevated temperatures induce thermal degradation of the
biomass components, which results in formation of cavities within the
cell.21–22,38 Second, the pore wall starts to collapse because water has
been removed. This will result in closure of the pore wall due to irre-
versible hydrogen bonding.39–40 Third, the anisotropic drying shrink-
age of cell wall layers results in drying stresses, which can damage
wood’s cell walls.41–43
Kauman explained that total cell wall collapse during drying
can be due to liquid tension collapse and stress collapse, which
lead to biomass shrinkage.44 It has been reported that the collapse
intensity increases with increased temperature during drying. 45–48
Borrega  and Kärenlampi stated that exposure of wood to tem-
peratures of >150°C during drying may cause thermal degradation
of its structure, often accompanied by loss of mass.38 The degree
of structural damage depends on tree species as well as process
parameters, such as duration of treatment, temperature, and relative
humidity.21–23 All structural changes observed in the biomass due
to moisture loss influence the biomass’s mechanical properties, like
bulk density and grindability.
Hemicellulose: Hemicellulose is a branched polymer (also a poly-
saccharide) that consists of shorter chains (500–3000 sugar units)
as opposed to the 7000–15 000 glucose molecules per unbranched
polymer observed in cellulose, as shown in Figure 2. Hemicellulose
accounts for about 25–35 wt% of biomass. The polysugars that
form hemicelluloses vary differently among woods and much more
among plants. In the case of deciduous wood, the hemicellulose
fraction is very different. Deciduous wood contains 80–90% of
4-O-methylglucoronoxylan (referred to as xylan), whereas conifer-
ous wood (most softwood) contains 60–70% of glucomannan and
15–30 wt% of arabinogalactan.18
Hemicellulose undergoes major decomposition reactions at
­t orrefaction temperatures of 200–300°C, resulting in different
100
90
Acid lignin
80
70
Weight (%)
60 Milled
wood lignin
50
40
Cellulose
30 Xylan
Wood
20
10
0 degree of polymerization in nature is difficult to measure since it is
100 200 300 400 500
Temperature (°C)
Figure 4. Thermogravimetry of cottonwood and its constituents35
© mary ann liebert, inc.
research
c­ ondensable and noncondensable products. Thermal degradation
of hemicellulose initiates at 150°C, with the majority of weight loss
occurring above 200°C, depending on the chemical nature of the
hemicellulose and the relationship with lignin within the cell.18,49
Compared to cellulose, hemicellulose generally evolves as light
volatiles, producing fewer tars and less char. Many researchers have
noted that major hemicellulose decomposition reactions occur at
temperatures between 220°C and 280°C.9 Native hemicellulose is
partly depolymerized by hydrolysis and/or thermal chain scission to
provide “reacting” hemicellulose. This intermediate is decomposed
by acid and radical reactions to yield many substances (e.g., furfu-
ral) that recombine to form torrefied hemicellulose. Water and acids
form during the above reactions and are released into the reaction
environment.50 Some of this water may be reused to depolymerize
hemicellulose or to release acids from the hemicellulose by hydro-
lysis of acetate groups. These radical reactions can also result in the
formation of acids.
Cellulose: Cellulose is a high-molecular-weight polymer (molecular
weight of 106 and higher) that makes up the fibers in wood and other
biomass. The multiple hydroxyl groups on the glucose residues from
one chain form hydrogen bonds with oxygen molecules on the same
or a neighboring chain. These chains are firmly held together side
by side, resulting in the formation of microfibrils with high tensile
strength. This strength is important in cell walls, where the micro-
fibrils are meshed into a carbohydrate matrix, conferring rigidity to
plant cells. Cellulose strands with hydrogen bonds (dashed within
and between cellulose molecules) are shown in Figure 2.
Cellulose degradation occurs between 240 and 350°C, resulting
in anhydrous cellulose and levoglucosan.18 The crystalline struc-
ture resists thermal depolymerization better than do unstructured
hemicelluloses. Amorphous regions in the cellulose contain waters
of hydration and hold free water within the plant. When heated
rapidly, this water is converted to steam, which can further rupture
the cellulose structure.
Cellulose also undergoes polymer restructuring similar to hemi-
cellulose through depolymerization reactions, but to a lesser extent.
The water and acids released due to hemicellulose depolymerization
can result in more cellulose degradation. This disordered cellulose is
thermostable and contains furan, aliphatic, and keto groups.
Lignin: Lignin is an amorphous, highly branched, cross-linked
macromolecular polyphenolic resin with no exact structure. Lignin
fills the spaces in the cell wall between cellulose, hemicellulose,
and pectin components. It is covalently linked to hemicellulose
and thereby cross-links different plant polysaccharides, conferring
mechanical strength to the cell wall and, by extension, to the plant
as a whole. It is relatively hydrophobic and aromatic in nature. The
fragmented during extraction and the molecule consists of various
types of substructures that appear to repeat in a haphazard manner.
An example of a lignin polymer is provided in Figure 2.
• Vol. 7 No. 5 • october 2011 INdustrial Biotechnology 389
research
Lignin decomposes when heated to 280–500°C, producing
­ henols via the cleavage of ether bonds and scissioning of ­carbon–
p During torrefaction, three different products are produced:
carbon bonds.18,49 Lignin is difficult to dehydrate and thus converts
to char more than cellulose or hemicellulose. Evans et al showed
that lignin releases some small phenolic fragments within the
­torrefaction temperature range.51 Grafting torrefied hemicellulose
onto cellulose and/or lignin may also be possible, indicating that
all of the biomass polymers can participate in torrefaction.9 The
biomass decomposition reactions at different temperature regimes
are given in Figure 5.28
char CO2+CO+H2 <250°C
Biomass
char+tar+CO2+CO+H20+H2+CH2+CnHm >400°C
H2+CH2+CnHm >700°C+residence time
Figure 5. Biomass decomposition reactions at different temperature
regimes28
Organic extractables: Organic extractables include fats, waxes,
alkaloids, proteins, phenolics, simple sugars, pectins, mucilages,
gums, resins, terpenes, starches, glycosides, saponins, and essential
oils.18 Again, these vary in amount and composition depending on
the type of biomass. Hence, devolatilization behavior depends on
the amount and location of the biomass extractables.
Phase
Solid
Biomass
Liquid (condensable)
Gas (permanent)
Figure 6. Products formed during torrefaction of biomass10
390 INdustrial Biotechnology october 2011
Torrefaction products
(1) brown to black uniform solid biomass, which is used for bioen-
ergy applications, (2) condensable volatile organic compounds com-
prising water, acetic acid, aldehydes, alcohols, and ketones, and (3)
noncondensable gases like CO2, CO, and small amounts of methane.
Release of these condensable and noncondensable products results
in changes in the physical, chemical, and storage properties of
biomass. Several studies have also investigated the physical prop-
erties and chemical composition of the liquids and gases released
during torrefaction.11,14,52 Bergman et al provide an overview of the
torrefaction products, based on their states at room temperature,
which can be solid, liquid, or gas, as shown in Figure 6.10 The solid
phase consists of a chaotic structure of the original sugars and
reaction products. The gas phase includes gases that are considered
permanent gases, and light aromatic components such as benzene
and toluene.
The condensables, or liquids, can be further divided into four
subgroups: (1)  reaction water produced from thermal decom-
position, (2) freely bound water that has been released through
evaporation, (3) organics (in liquid form), which consist of organics
produced during devolatilization and carbonization, and (4) lipids,
which contain compounds such as waxes and fatty acids.
Condensable and noncondensable products are emitted from
the biomass based on heating rate, torrefaction temperature and
time, and biomass composition. The emission profiles of these
products greatly depend on the moisture content in the biomass.
After torrefaction
Groups of components
• Original sugar structures
• Modified sugar structures
• Newly formed polymeric structures
• Char
• Ash
H2O
Organics:
Sugars, polysugars, acids, alcohols, furans, ketones
Lipids
Terpenes, phenols, fatty acids, waxes, tanins
• H2, CO, CO2, CH4
• CxHy, toluene, benzene
 research

At high moisture content, some of the surfactants present in the
biomass (oleic acid, fatty acids, and resin acids) help solubilize
the ­m onoterpenes in water and transport them through the
wood matrix. At low moisture content, the emission of different
compounds like terpenes is driven by vapor pressure and tem-
perature. During ­torrefaction the raw material loses most of its
moisture and other volatiles, which have a low heat value. Many
researchers have worked on identifying gas composition in terms
of quantity and quality.7,53 The type and amount of gas released
as off-gas ­during torrefaction depend on the raw material type
(i.e., anatomical, molecular, and chemical composition) and tor-
refaction process conditions, including the process temperature
and residence time.
Mass and energy yield
A typical mass-and-energy distribution is shown in Table 3,
which illustrates the preservation of mass and energy in the solid
product. One can see that besides the solids, mostly water is pro-
duced during torrefaction, and the energy content of the volatiles is
mainly preserved in the lipids and organics. Torrefaction operating
conditions and biomass properties have a significant impact on the
amount of solid residue remaining and the volatile and gaseous
products produced during torrefaction. Thus, solid, volatiles, and
Comparing the conversion of agricultural residues (i.e., rice straw
and rape stalk) to woody biomass from the literature, the solid-to-
liquid conversion of the former is much higher than that of the
latter under the same torrefaction conditions (i.e., temperature and
residence time).10 This is due to the higher volatile matter content
in the agricultural residues and the decomposition of hemicellu-
lose, the main fraction decomposed in the torrefaction temperature
range. Bridgeman et al drew similar conclusions where the mass
yield (dry ash-free) was 55.1, 61.5, and 72.0% for wheat straw, reed
canary grass, and willow, respectively, at 290°C for 30 min.12
Condensable products
Water is the main condensable product of torrefaction. It is
released during drying when moisture evaporates and during dehy-
dration reactions between organic molecules. Acetic acid is also
a condensable torrefaction product that mainly originates from
­acetoxy- and methoxy-groups present as side chains in xylose units
■ Gaseous product ■ Liquid product ■ Solid product
Willow Larch Straw
100
90
Product yield (wt%)

gases analysis and results from different studies are not exactly 80
the same. 70
60
Prins et al conducted several torrefaction mass balances experi-
50
ments.14 This data is reproduced in Figure 7, which shows a large
40
difference in product distribution for different types of biomass. 30
Larch has the largest yield of solid product and smallest gas and 20
liquid yield, straw has higher gas and liquid fractions than woody 10
biomass, and willow is between woody biomass and straw. It was 0
n) n) n) n) n) n) n) n) n) n)
also found that an increase in torrefaction temperature leads to a
0 mi 0 mi 0 mi 0 mi 5 mi 0 mi 0 mi 0 mi 5 mi 0 mi
decrease in solid char yield and an increase in the volatile fractions, (5 10 (3 (6 (1 (3 (5 (3 (1 (3
0 °C °C ( 0°C 0°C 0°C 0°C 0°C 0°C 0°C 0°C
including liquid and noncondensable gases, which is a result of the 23 230 25 25 27 27 23 25 27 25
competition between charring and devolatilization reactions that
become more reactive at higher temperatures. Figure 7. Overall mass balance of several torrefaction experiments14

Table 3. Mass and energy distribution of willow at 280°C

for 17.5 min reaction time31
Condensable volatile yield (wt%)

14
Energy yield ■ Acetic acid ■ Lactic acid
12 ■ Water ■ Furfural
Reaction products Mass yield (lower heating value ■ Formic acid ■ Hydroxyacetone
10 ■ Methanol ■ Phenol
[LHV], daf) (%)
8
Solid 87.5 94.9
6
Lipids 1.4 3.4
4
Organics 1.7 1.6
2
Gases 1.4 0.1
0
Water 8.0 0.0 230°C 250°C 270°C 280°C 300°C
(50 min) (30 min) (15 min) (10 min) (10 min)
Note: (daf: dry and ash-free basis)
Figure 8. Condensable volatiles yields during torrefaction14

© mary ann liebert, inc. • Vol. 7 No. 5 • october 2011 INdustrial Biotechnology 391
research

(part of the hemicellulose fraction). Prins et al proved that smaller
quantities of formic acid, lactic acid, furfural, hydroxyl acetone,
and traces of phenol are also present in the volatile component.14
Yields of most of the condensable volatiles will increase with the
torrefaction temperature, as is shown in Figure 8. As a result, more
energy is transferred to the volatiles in the form of combustibles,
such as methanol and acetic acid.
Noncondensable products
The major gases formed during torrefaction are carbon diox-
ide and carbon monoxide with traces of other gases, as is shown
in Figure 9. 10,54 Carbon monoxide is the main source of the
noncondensable CV of torrefaction products.
The formation of CO2 may be explained by decarboxylation of
acid groups in wood and other herbaceous biomass. The formation
of CO cannot be explained by dehydration or decarboxylation reac-
tions. The increased CO formation is reported to be caused by the
reaction of carbon dioxide and steam with porous char.55 Traces
of hydrogen and methane are also detected in noncondensable
products. A gas composition comparison between wood and agri-
cultural residues indicates that the latter is characterized by higher
100
■ Mass composition
■ Energy composition
75
Percent
CO2 production.11,14,52 In addition, a kinetic study on the generation
of noncondensable gases shows that the gases are formed through
parallel independent first-order reactions.13
The composition of the noncondensable volatiles produced dur-
ing different torrefaction conditions is shown in Figure 10. The
ratio of CO to CO2 increased with temperature because cellulose and
lignin decompose at higher temperatures.14
Figure 11 shows the gas composition of the noncondensable
products over time, which do not total 100% because only the main
components (CO2 and CO) are shown.14 These results were obtained
by torrefaction of larch and willow at 250°C for 5, 15, and 30 min.
It was found that the ratio of CO2 to CO decreases with time, in line
with the theory that CO is formed in a secondary reaction.14
Torrefaction process efficiency
The torrefaction process increases the energy density and
improves the overall fuel properties of the torrefied product.
However, torrefaction is an energy-intensive pretreatment operation
due to the high temperatures used in the process. Reusing the excess
heat generated in the process will have a significant impact on the
overall efficiency of the system. Also, both the condensable and
noncondensable volatiles have a significant impact on the energy
efficiency of the process. To achieve high energy efficiency at low
cost, innovative torrefaction technologies have been developed to
capture excess heat and volatile energy; the Energy Research Center
of the Netherlands’ (ECN’s) torrefaction process is one of the most

50 mature, as is shown in Figure 12.10

The depicted process layout is based on the direct heating of
biomass during torrefaction by means of recycled hot gas. The
25
hot gas consists of the torrefaction gas and is re-pressurized to
compensate for the pressure drop in the recycle loop. It heats the
0 recycled gas to deliver the required heat demand in the torrefaction
CO CO2 Other
reactor. Combustion of the liberated torrefaction gas produces the
Figure 9. Main permanent gases compositions10,54 necessary heat for torrefaction and pre-drying. A utility fuel can be
used when the energy content of the torrefaction gas is insufficient

■ Carbon monoxide ■ Carbon dioxide

Noncondensable volatile yield (wt%)

3.5
Larch Straw Willow Larch
3.0
Composition of non-condensable

100
2.5 ■ Carbon monoxide 90
■ Carbon dioxide 80
gas phase (vol %)

2.0
70
1.5 60
50
1.0
40
0.5 30
0.0 20
230°C 250°C 270°C 290°C 250°C 10
(50 min) (30 min) (15 min) (10 min) (30 min) 0
5 min 15 min 30 min 5 min 15 min 30 min

Figure 11. Change in gas composition over time for torrefaction at

Figure 10. Noncondensable volatiles yields for larch and straw14 250°C 14

392 INdustrial Biotechnology october 2011

 research

to thermally balance the torrefaction process and to provide stabil- content of about 10% wt, which makes up a total of 60% wt of
ity and control of the combustion process. Bergman et al identified incombustible components of the total emissions. The exact amount
this process concept as the most promising for torrefaction, which is determined by specific conditions and by the moisture content of
achieves autothermal operation when the total heat demand of the the biomass feed.
process (drying and torrefaction) is balanced by the energy content Bergman et al10,11 and Bergman32 further examined the CV of the
of the torrefaction gas.10 The torrefaction conditions (temperature torrefaction gas experimentally, while mass-and-­energy balance
and residence time) are the crucial variables to tune the thermal thermal-process efficiency, autothermal operation, and combustibil-
balance (i.e., the energy yield of torrefaction and hence the energy ity of the torrefaction gas were investigated by means of process sim-
content of the torrefaction gas). ulations. In their studies, the yield of reaction water varied between
The torrefaction gas consists of a wide variety of combustible 5 and 15% wt, resulting in a concentration of 50–80% wt in the tor-
organic components. However, the main components of torrefaction refaction gas (excluding free water from the feedstock). The reaction
gas are incombustible (water and CO2). The given product distribu- water yield increased with residence time and temperature, while its
tion in Figures 6–8 shows that the torrefaction gas is rather wet. concentration decreased. Consequently, the relative contribution of
Even when completely dry biomass is torrefied, the torrefaction combustible products increases with increased temperature and resi-
gas is expected to have a water content of over 50% wt and a CO2 dence time, as does the CV, which ranges from 5.3–16.2 ­MJ/­Nm3.
Despite the high water content of the torrefaction
gas, the CV is relatively high. It can be compared to
Air
Utility fuel producer gas from air-blown biomass gasification
Flue gas (4–7 MJ/Nm3) and to syngas from an indirectly ­heated
gasification process (15–20 MJ/Nm3). Based on this
Torrefaction Combustion comparison, the torrefaction gas should be combus-
gases Gas
Recycle Torrefied tible and can play an important role in the torrefac-
biomass tion process.10,11,32 Typical experimental results for
Biomass Drying Torrefaction Cooling torrefaction mass and energy yields and gas-phase
composition for willow are given in Figure 13.
DP
Solid torrefied biomass properties
Torrefaction of biomass significantly changes its
Flue gas Heat exchange physical and chemical properties like moisture con-
Flue gas
tent, density, grindability, pelletability, hydrophobic-
ity, CV, proximate and ultimate composition, and
Figure 12. The envisaged conceptual structure of ECN’s torrefaction process10
storage behaviors in terms of off-gassing, spontane-
ous combustion, and self heating.15

Gas phase composition Physical properties

CO = 0.1% Moisture content
CO2 = 3.3% Normally, feedstock moisture content ranges
H2O = 89.3% from 10–50%, but because torrefaction is a deep
Acetic acid = 4.8% drying process, moisture content is reduced to
Furfural = 0.2% 1–3% on a weight basis, depending on the torrefac-
Methanol = 1.2% tion conditions. 31 Typically, torrefaction achieves
Formic acid = 0.1%
an equilibrium moisture content of 3% and a
Remainder = 1.0%
Torrefaction ­reduction of mass by 20–30% (primarily by release
(32 min, 260°C) of water, carbon oxides, and volatiles), while
retaining 80–90% of the wood’s original energy
Feed: Willow Torrefied willow content.9 Reduction in moisture during torrefaction
Size: 10–30 mm Size: 10–30 mm provides three main benefits: (1) reduced moisture
LHV = 14.8 MJ/kg LHV = 18.5 MJ/kg level for the conversion process, (2) reduced trans-
Mass yield: 75.3%
MC = 14.4% (w.b.) MC = 1.9% (w.b.) portation costs associated with ­moving unwanted
Energy yield: 94.3%
Fixed carbon: 16.8% Fixed carbon: 22.1% water, and (3) the prevention of biomass decompo-
sition and moisture absorption during storage and
Figure 13. Typical experimental results of torrefaction of willow54 ­transportation.

© mary ann liebert, inc. • Vol. 7 No. 5 • october 2011 INdustrial Biotechnology 393
research

Table 4. Bulk density (dry basis) of Eucalyptus grandis wood which is mainly associated with the breakdown of the ­hemicellulose
in three different treatments56 matrix and depolymerization of the cellulose, ­resulting in decreased
fiber length.10,31 Particle length is also decreased, but not the ­diameter
Treatment Bulk density (g/cm3) Percentage loss
per se, resulting in better grindability, handling characteristics, and
Control* 0.85 a
— flowability through processing and transportation systems. Also
220°C 0.83 a
2.35 during the torrefaction process, the biomass tends to shrink, become
lightweight, flaky, and fragile, and lose its mechanical strength,
250°C 0.79b
7.06
making it easier to grind and pulverize.1
280°C 0.73 c
14.12 Bergman and Kiel conducted studies on the energy requirements
Note: Means followed by the same letter are statistically similar at the 5% for grinding raw and torrefied biomass like willow, woodcuttings,
probability level. demolition wood, and coal using a heavy-duty cutting mill.31 They
*Average moisture content of control treatment = 15%. concluded that power consumption reduces dramatically when bio-
mass is first torrefied. This reduction ranges from 70–90%, based on
the conditions under which the material is torrefied. Bergman and
Bulk and energy density Kiel also found that the capacity of the mill increases by a factor
Mass loss in the form of solids, liquids, and gases during of 7.5–15%.31 The most important observation they made was that
t­orrefaction cause the biomass to become more porous. This results the size-reduction characteristics of torrefied biomass were similar
in significantly reduced volumetric density, typically between to coal. Phanphanich and Sudhagar also observed a decrease in
180  and 300 kg/m 3, depending on initial biomass density and the grinding energy from 237.7–37.6 kWh/t, from raw to torrefied
­torrefaction conditions.31 Oliveira-Rodrigues and Rousset’s study on ­forest biomass at 300°C for 30 min.57 Figure 14 gives the typical
the effect of torrefaction on energy properties of Eucalyptus gran- grinding energy reduction for torrefied wood.58
dis wood indicated a bulk density loss of 14.12% when the wood
was torrefied at 280°C for 30 min.56 Table 4 shows the loss in bulk Particle size distribution, sphericity, and particle surface area
density for torrefaction temperatures ranging from 220–280°C.56 In Particle-size distribution curves, sphericity, and surface area are
spite of reduced bulk density, the energy density increases. Bergman important parameters for understanding flowability and combustion
et al observed that the calorific density of pine woodchips increased behavior during cofiring. Many researchers observed that ground,
from 11–20 MJ/kg.11 Many researchers have observed that the torrefied biomass produced narrower, more uniform particle sizes
energy density of both lignocellulosic and woody biomass increased compared to untreated biomass due to its brittle nature, which is
significantly after torrefaction (>60%). similar to coal. Phanphanich and Sudhagar studied torrefied pine
chips and logging residues and found that smaller particle sizes
Grindability are produced compared to untreated biomass.57 In addition, they
Biomass is highly fibrous and tenacious in nature; fibers form observed that the particle distribution curve was skewed towards
links between particles and make handling the raw ground samples smaller particle sizes with increased torrefaction temperatures.57
difficult. During torrefaction, the biomass loses its tenacious nature, Torrefaction also significantly influences the sphericity and
particle surface area. Phanphanich and Sudhagar’s results also
indicated that sphericity and particle surface area increased as the
900 torrefaction temperature was increased up to 300°C.57 For ground,
800 ■ 220°C torrefied chips, they found that the sphericity increased from 0.48%
Grinding energy (kWh/t)

700 ■ 240°C
■ 260°C to 0.62%, concluding that an increase in particle surface area or
600 decrease in particle size of torrefied biomass can be desirable prop-
500 erties for efficient cofiring and combustion applications.57 Also, the
400 bulk and particle densities of ground torrefied biomass increase
300 because the inter- and intraparticle voids generated after milling are
200 reduced.59–60 Research studies have indicated that ground torrefied
100 material results in a powder with favorable size distribution and
0 spherical particles, allowing torrefied powder to meet the smooth
Untreated 5 min 20 min 40 min 60 min
fluidization regime required for feeding it to entrained-flow pro-
Duration (min)
cesses (i.e., gasifier and pulverized coal).

Figure 14. Grinding energy of beech as a function of ­torrefaction Pelletability

duration, obtained with a Retsch ZM1 ultracentrifugal mill equipped Variability in feedstock quality due to differences in the types
with a 500‑mm grid58 of raw materials, tree species, climatic and seasonal variations,

394 INdustrial Biotechnology october 2011

 research

Drying Torrefaction Size reduction Densification Cooling TOP pellets

Figure 15. Flow diagram for production of torrefied wood pellets32

1.8
storage conditions, and time significantly influence the quality of Willow C Untreated willow
Willow B (240°C:10 min)
biopellets.61 Torrefying the biomass before pelletization, however, 1.6 (240°C:60 min) Untreated
Willow A miscanthus
produces uniform feedstock with consistent quality. 1.4
(290°C:10 min)
Willow D Miscanthus C
Lignin in the biomass is considered to be the basic binding agent; (290°C:60 min) (240°C:10 min)

Atomic H: C ratio
1.2 Miscanthus B
(240°C:60 min)
thus, the pelletability of any biomass is evaluated based on lignin Miscanthus D
Miscanthus A
content. In general, higher amounts of lignin improve binding and 1.0 Bituminous
(290°C:60 min)
(290°C:10 min)
coal Lignite
reduce the severity of process conditions. The torrefaction process 0.8
opens more lignin‑active sites by breaking down the hemicellulose
0.6
matrix and forming fatty unsaturated structures, which creates Charcoal

better binding. Densification following torrefaction has been con- 0.4 Anthracite

sidered by several researchers.9,11,32,62–63 These studies indicate that, 0.2

compared to raw biomass pelletization, the required pressure and
energy consumption can be reduced by a factor of 2 when the mate-
rial is densified at a temperature of 225°C. In addition, densification
tests on torrefied biomass using a Pronto-Press indicated that the
density of torrefied and pelletized material varied in the range of
750–850 kg/m3 compared to conventional wood pellets, which are
in the range of 520–640 kg/m3. The energy density and CV have
also increased by approximately 70–80%, which is comparable to
sub-bituminous coal.31 Figure 15 shows the flow diagram for the
TOP process that Bergman proposed.32
Chemical compositional changes
Besides improving physical attributes, torrefaction also results in
significant changes in proximate and ultimate composition of the
biomass and makes it more suitable for fuel applications. Sadaka
and Negi’s15 study on torrefaction of wheat straw, rice straw, and
cotton gin waste at 200, 260, and 315°C for 60, 120, and 180 min,
respectively, concluded that moisture content was reduced at the
conditions (260°C for 120 min) for all three feedstocks by 70.5, 49.4,
and 48.6%, and the heating value increased by 15.3, 16.9, and 6.3%,
respectively. Zanzi et al, in their study on miscanthus torrefaction,
made similar observations, where increasing temperature from 230
to 280°C and time from 1–3 h increased the carbon content and
decreased the hydrogen, nitrogen, and oxygen content.7 At 280°C,
the carbon content increased to about 52% from an initial value of
43.5%, while hydrogen and nitrogen content decreased from 6.49
to 5.54% and 0.90 to 0.65%, respectively, for 2 h of torrefaction.
In general, increased torrefaction temperatures result in increased
carbon content and decreased hydrogen and oxygen content due to
the formation of water, CO, and CO2.
This process also causes the hydrogen-to-carbon (H/C) and
oxygen-to-carbon (O/C) ratios to decrease with increasing tor-
refaction temperature and time, which results in less smoke and
water-vapor formation and reduced energy loss during combustion
0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
Atomic O: C ratio
Figure 16. Van Krevelen diagram for torrefied willow and
miscanthus64
and gasification processes. In torrefaction studies of reed canary
grass and wheat straw torrefaction at 230, 250, 270, and 290°C for
30 min residence times, Bridgeman et al found that the moisture
content decreases from an initial value of 4.7–0.8%.12 They found
that carbon increased 48.6–54.3%, while the hydrogen and nitrogen
content decreased from 6.8–6.1% and 0.3–0.1%, respectively. In
another study, Bridgeman et al made a Van Krevelen diagram for
torrefied willow and miscanthus, as is shown in Figure 16.64 It is
clear that at higher temperatures and residence times, the atomic
O/C and H/C ratios are closer to that of lignite coal. Table 5 shows
the effect of different torrefaction temperatures on ultimate compo-
sitional changes in woody and herbaceous biomass.12 Table 6 shows
the physiochemical composition of pine and heat-treated pine at
torrefaction temperatures from 240–290°C and indicates that as tor-
refaction temperature increases, fixed carbon increases and volatiles
and moisture content decrease.65
Calorific value
Biomass loses relatively more oxygen and hydrogen than ­carbon
during torrefaction, which increases the calorific value of the prod-
uct.6 The net CV of torrefied biomass is 18–23 MJ/kg (lower heating
value [LHV], dry) or 20–24 MJ/kg (higher heating value [HHV],
dry).11,66 The mass and energy in the torrefied biomass is preserved
in the solid product for a long time, as the material does not degrade
with time. Table 5 provides the ultimate analysis and elemental
composition of torrefied woody and herbaceous biomass proper-

© mary ann liebert, inc. • Vol. 7 No. 5 • october 2011 INdustrial Biotechnology 395
research

Table 5. Ultimate analysis, moisture content, and HHV (dry ash free basis) of untreated and torrefied biomass12

Raw Torrefaction temperature

Reed canary grass 0°K 503°K 523°K 543°K 563°K

C (%) 48.6 49.3 50.3 52.2 54.3

H (%) 6.8 6.5 6.3 6.0 6.1

N (%) 0.3 0.1 0.0 0.1 0.1

O (%) 37.3 N/A 37.0 37.3 36.3

Moisture (%) 4.7 2.5 1.9 1.3 1.2

CV (kJ/kg) 19,500 N/A 20 000 20 800 21 800

Wheat straw

C (%) 47.3 48.7 49.6 51.9 5.6

H (%) 6.8 6.3 6.1 5.9 1.0

N (%) 0.8 0.7 0.9 0.8 27.6

O (%) 37.7 N/A 35.6 33.2 0.8

Moisture (%) 4.1 1.5 0.9 0.3 0.8

CV (kJ/kg) 18,900 19 400 19 800 20 700 22 600

Willow

C (%) 49.9 50.7 51.7 53.4 54.7

H (%) 6.5 6.2 6.1 6.1 6.0

N (%) 0.2 0.2 0.2 0.2 0.1

O (%) 39.9 39.5 38.7 37.2 36.4

Moisture (%) 2.8 0.5 0.1 0.1 0.0

CV (kJ/kg) 20,000 20 600 20 600 21 400 21 900

Table 6. Physiochemical analysisa of pine and heat-treated pine65

T (°C)b pc 240 250 260 270 290

Fixed carbon, % 20.64 23.55 25.59 25.69 29.38 35.39

Elementary analysis
C, % 50.98 51.14 51.93 53.78 53.57 58.08
O, % 42.80 42.70 42.18 40.66 40.67 36.40

Pentosans, % 9.61 5.93 5.90 3.10 2.54 1.40

Lignin, % 22.84 24.90 28.72 33.44 39.23 53.47

Extractables, % d
14.67 8.19 14.09 19.35 16.49 17.98

Moisture, % e
10.80 5.66 4.08 3.96 3.76 3.88

Yield, % — 86.2 81.8 75.7 66.4 48.8

a
In each case, the mean result given was obtained from a minimum of four different experiments. bTorrefaction time: 30 min. cNative pine. dNeutral-solvent extractables
(ethanol, benzene, boiling water). ePowder samples left at the laboratory atmosphere still had a constant humidity.

396 INdustrial Biotechnology october 2011


ties as compared to raw materials. Sadaka and Negi also observed
that the highest heating value of 22.75 MJ/kg, or 9761 BTU/lb, was
achievable at torrefaction conditions of 315°C and 3 h.15
Storage aspects of torrefied biomass
Off-gassing
Storage issues like off-gassing and self-heating may also be
insignificant in torrefied biomass, as most of the solid, liquid, and
gaseous products that are chemically and microbiologically active
are removed during the torrefaction process. Kuang et al’s67 and
Tumuluru et al’s 68 studies on wood pellets concluded that high stor-
age temperatures of 50°C can result in high CO and CO2 emissions,
and the concentrations of these off-gases can reach up to 1.5% and
6% for a 60-day storage period. These emissions were also found to
be sensitive to relative humidity and product moisture content. The
same researchers at the University of British Columbia conducted
studies on off-gassing from torrefied woodchips and indicated that
CO and CO2 emissions were very low — nearly one-third that of
regular woodchips at room temperature (20°C). This could be due
to low moisture content and reduced volatile content, which could
result in less reactivity with the storage environment.
Hydrophobicity
In general, the uptake of water by raw biomass is due to the pres-
ence of OH groups. Torrefaction produces a hydrophobic product by
destroying OH groups and causing the biomass to lose the capacity
to form hydrogen bonds.69 Due to these chemical rearrangement
reactions, nonpolar unsaturated structures are formed, which pre-
serve the biomass for a long time without biological degradation,
similar to coal.31,70
Bergman determined the hydrophobicity of torrefied pellets by
immersing them in water for 15 h.32 Their hydrophobic nature was
evaluated based on the state of the pellet after this period and by
20
16
Moisture uptake (% mass)
decrease
12
8
◆ Control
4 ■ 250°C
▲ 300°C
0 be a good replacement for wood pellets in cofiring and gasification
0 20 40
Exposure time (h)
Figure 17. Moisture uptake by the torrefied wood pellets made from
a 0.8 mm particle size71
© mary ann liebert, inc.
research
gravimetric measurement to determine the degree of water uptake.
Bergman’s study indicated that raw pellets swelled rapidly and dis-
integrated into original particles.32 Torrefied pellets produced under
optimal conditions, however, did not disintegrate and showed little
water uptake (7–20% on a mass basis). Bergman also concluded that
torrefaction conditions play a vital role in the hydrophobic nature
of biomass.32 Sokhansanj et al compared the moisture uptake of
the torrefied biomass to the untreated biomass and found a 25%
decrease in water uptake when compared to the control, as shown
in Figure 17.71
Biomass is porous, often moist, and prone to off-gassing and
self-heating due to chemical oxidation and microbiological activity.
In general, the biomass moisture content plays an important role in
initiating chemical and microbial reactions. Moisture content cou-
pled with high storage temperatures can cause severe off-­gassing
and self-heating from biomass-based fuels. Another important stor-
age issue for ground torrefied biomass is its reactivity in powder
form, which can result in fire during storage. It is preferred to store
the torrefied biomass in an inert environment to avoid incidents of
spontaneous combustion. In his laboratory-scale combustion stud-
ies of torrefied wood, Kiel found that it is highly reactive, similar
to coal.53
Table 7 shows the product characteristics of torrefied biomass
compared to coal and wood, revealing that torrefied wood closely
resembles charcoal. The major difference between charcoal and
­torrefied wood is the volatile content. Volatiles are lost during
charcoal production, which also means a possible loss of energy.72
On the other hand, during torrefaction, most of the volatiles are
retained. It is also recommended that every form of carbonization
be avoided during torrefaction. As Table 7 shows, torrefied pellets
have product characteristics, like handling, milling, and transport
requirements, similar to coal.65 Torrefied pellets allow for higher
cofiring percentages up to 40% due to matching fuel properties
with coal, and they can use the existing equipment setup for coal.
Torrefied biomass applications
Pulverized fuel combustion in coal-fired power stations and
25%
entrained-flow gasification are particularly interesting product
outlets for biomass. In both applications, biomass has to be fed
to the reactor as a powder, which is difficult, costly, and achiev-
able only at very low capacity in classical coal mills. Due to this
limitation, wood pellets are currently the state-of-the-art for
cofiring, as they consist of sufficiently small particles. Consequently,
wood pellets also have some limitations in terms of energy content
and moisture uptake during storage and transportation. Torrefied
biomass, because it is energy-dense and hydrophobic in nature, can
0 plants. The high fuel quality of torrefied biomass makes it very
attractive for combustion and gasification applications. Due to high
calorific values, the thermal energies of the combustion and gas-
ification system can be improved significantly.31 However, data is
lacking on milling, handling, storing, transporting, and combusting
• Vol. 7 No. 5 • october 2011 INdustrial Biotechnology 397
research

Table 7. Indicative properties of different biomass and coal-based fuels72

Wood Wood pellets Torrefied pellets Charcoal Coal

Moisture content (% wt) 30–45 7–10 1–5 1–5 10–15

Calorific value (MJ/kg) 9–12 15–16 20–24 30–32 23–28

Volatiles (% db) 70–75 70–75 55–65 10–12 15–30

Fixed carbon (% bd) 20–25 20–25 28–35 85–87 50–55

Bulk density (kg/l) 0.2–0.25 0.55–0.75 0.75–0.85 ~0.2 0.8–0.85

Volumetric energy density (GJ/m3) 2.0–3.0 7.5–10.4 15.0–18.7 6.0–6.4 18.4–23.8

Dust Average Limited Limited High Limited

Hygroscopic properties Hydrophilic Hydrophilic Hydrophobic Hydrophobic Hydrophobic

Biological degradation Yes Yes No No No

Milling requirement Special Special Classic Classic Classic

Handling requirements Special Easy Easy Easy Easy

Product consistency Limited High High High High

Transport cost High Average Low Average Low

large amounts of torrefied biomass at a commercial scale. Common
torrefied biomass applications include: (1) high-quality smokeless
solid fuels for industrial, commercial, and domestic applications,
(2) solid fuel for cofiring directly with pulverized coal at electric
power plants, (3) an upgraded feedstock for fuel pellets, briquettes,
and other densified biomass fuels, and (4) high-quality biomass
solid fuel for advanced bioenergy application.
Conclusions
Interest in research on torrefaction of biomass materials is grow-
ing. Its potential to improve the quality of both herbaceous and
woody materials provides a path for using these resources in
many energy applications. This review has provided insight into
the ­limitations of raw biomass materials to meet the needs of
energy providers and has discussed options for improving biomass
­resources for use in power and liquid-fuel production. Based on this
review, nine areas of discussion are summarized below. These areas
provide a brief understanding of the potential use of torrefaction as
a means of improving the physical, chemical, and rheological char-
acteristics of biomass materials. As with most biomass-upgrading
options, however, torrefaction, and the subsequent densification
processes, need to be evaluated for energy inputs to understand the
logistical advantages and end use of the product.
1. Torrefied biomass, in general, defines a group of products
resulting from the partially controlled and isothermal pyrolysis
of biomass occurring at the 200–300°C temperature range.
2. The most common torrefaction reactions include (a) devolatil-
ization and carbonization of hemicelluloses, and (b) depoly-
merization and devolatilization of lignin and cellulose.
3. T orrefaction of the biomass helps in developing a uniform
feedstock with minimum variability in moisture content.
4. Torrefaction of biomass improves (a) energy density, grindability,
and pelletability index ratings, (b) ultimate and proximate com-
position by increasing the carbon content and CV and decreas-
ing the moisture and oxygen content, and (c) biochemical com-
position by decomposing the hemicelluloses and softening the
lignin, which results in better binding during pelletization.
5. Ground torrefied biomass has improved sphericity, particle
surface area, and particle size distribution.
6. Densification of torrefied material reduces specific energy
consumption and increases throughput by about a factor of 2
compared to raw or untreated biomass.
7. During torrefaction the biomass loses most of the low energy
content of the material, like (a) solids, which include original
sugar structures and other newly formed polymeric structures,
and (b) liquids, which include condensables, like water, organ-
ics, and lipids, and (c) gases, which include H2, CO, CO2 and
CH4, CxHy, toluene, and benzene.
8. Torrefaction preserves the biomass for a long time without
biological degradation due to the chemical rearrangement
reactions and formation of nonpolar unsaturated structures.
9. Torrefied biomass has been successfully used as an upgraded
solid fuel in electric power plants and gasification plants.
Not all aspects of torrefaction and its influence on other pro-
cessing operations have been explored. Opportunities for future
research can include:
a. Optimizing torrefaction processes based on activation energies
required to degrade the cellulose, hemicelluloses, and lignin.

398 INdustrial Biotechnology october 2011


b. U  nderstanding the torrefaction process at a molecular level by
identifying different functional groups and energies associated
with breaking the bonds.
c. Understanding the spontaneous torrefaction process reactions
using Fourier transform infrared (FTIR) and Raman spectros-
copy.
d. Understanding the severity of the torrefaction process based on
color changes using the Hunter colorimeter.
e. Studies on thermogravimetrics to establish the weight-loss
kinetics.
f. Studies on microstructural changes in biomass at different
temperature regimes.
g. Testing integrated processes such as torrefaction and densifi-
cation.
h. Calculating energies associated with condensable and noncon-
densable products produced during torrefaction and the ability
to reuse them to improve the overall process efficiency.
i. Understanding the storage behavior of torrefied biomass in
terms of off-gassing and spontaneous combustion at different
storage temperatures.
AC K N OW L E D G M E N T S
The authors acknowledge Leslie Park Ovard for her valu-
able contribution to the manuscript, and Gordon Holt, Allen
Haroldsen, Lisa Plaster, and Huan-Chen Tang, from Idaho National
Laboratory’s R&D Publications Support Team for their editorial
and graphics creation assistance. This work is supported by the
US Department of Energy, under DOE Idaho Operations Office
Contract DE-AC07-05ID14517. Accordingly, the US Government
retains and the publisher, by accepting the article for publication,
acknowledges that the US Government retains a nonexclusive,
paid-up, irrevocable, worldwide license to publish or reproduce
the published form of this manuscript, or allow others to do so,
for US Government purposes.
US DEPARTMENT OF ENERGY DISCLAIMER
This information was prepared as an account of work spon-
sored by an agency of the US government. Neither the US govern-
ment nor any agency thereof, nor any of their employees, makes
any warranty, express or implied, or assumes any legal liability
or responsibility for the accuracy, completeness, or usefulness
of any information, apparatus, product, or process disclosed, or
represents that its use would not infringe privately owned rights.
Any reference herein to any specific commercial product, process,
or service by trade name, trademark, manufacturer, or otherwise,
does not necessarily constitute or imply its endorsement, recom-
mendation, or favoring by the US Government or any agency
thereof. The views and opinions of the authors expressed herein
do not necessarily state or reflect those of the US Government or
any agency thereof.
© mary ann liebert, inc.
research
REFERENCES
1. Arias BR, Pevida CG, Fermoso JD, Plaza MG, Rubiera FG, and Pis-Martinez JJ.
Influence of torrefaction on the grindability and reactivity of woody biomass.
Fuel Process Technol 89(2), 169–175 (2008).
2. United Nations (UN). Kyoto protocol to the United Nations framework conven-
tion on climate change [Online]. Available at: http://unfccc.int/resource/docs/
convkp/kpeng.pdf [August 3, 2011] (1998).
3. Evergreen Renewables, LLC. Biomass torrefaction as a preprocessing step
for thermal conversion: Reducing costs in the biomass supply chain [Online].
Available at: http://evergreenrenewable.com/welcome_files/Biomass%20torre-
faction.pdf [August 3, 2011] (2009).
4. Pipatmanomai S. Overview and experiences of biomass fludized bed gasification
in Thailand. J Sustainable Energy Environ Special Issue, 29-11 (2011).
5. Sierra R, Smith A, Granda C, and Holtzapple MT. Producing fuels and chemi-
cals from lignocellulosic biomass. Chemical Engineering Process, Society of
Biological Engineering Special Section: Biofuels 104, S10–S18 [Online]. Available
at: http://www.aiche.org/uploadedFiles/SBE/MemberCenter/0808S10(2).pdf
[August 3, 2011] (2008).
6. Uslu A, Faaij A, and Bergman PCA. Pre-treatment technologies, and their effect
on international bioenergy supply chain logistics: Techno-economic evalua-
tion of torrefaction, fast pyrolysis, and pelletisation. Energy 33(8), 1206–1223
(2008).
7. Zanzi R, Ferro DT, Torres A, Soler PB, and Bjornbom E. Biomass torrefaction.
In: The 6th Asia-Pacific International Symposium on Combustion and Energy
Utilization. Kuala Lumpur, Malaysia, 20–22 May 2002 (2002).
8. Bourgeois JP and Doat J. Torrefied wood from temperate and tropical species:
Advantages and prospects. In: Bioenergy 84, 3, 153–159. Egneus H, Ellegard
A, Okeefe P, and Kristofferson L (eds), Elsevier Applied Science, London, United
Kingdom. (1984).
9. Lipinsky ES, Arcate JR, and Reed TB. Enhanced wood fuels via torrefaction.
Fuel Chemistry Division Preprints, 47(1), 408–410 (2002).
10. Bergman PCA, Boersma AR, Zwart RWH, and Kiel JHA. Torrefaction for biomass
co-firing in existing coal-fired power stations. Report ECN-C--05-013, ECN,
Petten (2005a).
11. Bergman PCA, Boersma AR, Kiel JHA, Prins MJ, Ptasinski KJ, and Janssen FGGJ.
Torrefied biomass for entrained-flow gasification of biomass. Report ECN-C--
05-026, ECN, Petten (2005b).
12. Bridgeman TG, Jones JM, Shield I, and Williams PT. Torrefaction of reed canary
grass, wheat straw, and willow to enhance solid fuel qualities and combustion
properties. Fuel 87(6), 844–856 (2008).
13. Prins MJ, Ptasinski KJ, and Janssen FGGJ. Torrefaction of wood: Part 1, Weight
loss kinetics. J Anal Appl Pyrolysis 77(1), 28–34 (2006a).
14. Prins MJ, Ptasinski KJ, and Janssen FGGJ. Torrefaction of wood: Part 2, Analysis
of products. J Anal Appl Pyrolysis 77(1), 35–40 (2006b).
15. Sadaka S and Negi S. Improvements of biomass physical and thermochemical
characteristics via torrefaction process. Environmental Progress Sustainable
Energy 28(3), 427–434. doi: 10.1002/ep.10392 (2009).
16. Moire L, Schmutz A, Buchala A, Yan B, Stark RE, and Ryser U. Glycerol is a
suberin monomer: New experimental evidence for an old hypothesis. Plant
Physiol 1999 March, 119(3), 1137–1146 (1999).
17. Esau K. Plant Anatomy, 2nd ed. John Wiley & Sons, New York, New York (1964).
18. Mohan D, Pittman CU, and Steele PH. Pyrolysis of wood/biomass for bio-oil:
A critical review. Energ Fuel 20(3), 848–889 (2006).
19. Shafizadeh F. Pyrolytic reactions and products of biomass. In: Fundamentals
of Biomass Thermochemical Conversion. 183–217. Overend RP, Milne TA, and
Mudge LK (eds), Elsevier Applied Science, London, United Kingdom (1985).
20. Williams PT and Besler S. The influence of temperature and heating rate on the
slow pyrolysis of biomass. Renewable Energy 7(3), 233–250 (1996).
• Vol. 7 No. 5 • october 2011 INdustrial Biotechnology 399
research
21. Stamm AJ. Thermal degradation of wood and cellulose. Ind Eng Chem 48,
413–417 (1956).
22. Esteves B, Marques AV, Domingos I, and Pereira H. Influence of steam heating
on the properties of pine (Pinus pinaster) and eucalypt (Eucalyptus globulus)
wood. Wood Sci Technol 41(3), 193–207 (2007).
23. Borrega M and Kärenlampi PP. Effect of relative humidity on thermal degrada-
tion of Norway spruce (Picea abies) wood. J Wood Sci 54, 323–328 (2008).
24. Sterling C. Effect of moisture and high temperature on cell walls in plant tis-
sues. Food Res 20(5), 474–479 (1955).
25. Kärenlampi, PP, Tynjala P, and Strom P. Phase transformation of wood cell wall
water. J Wood Sci 51, 118–123 (2005).
26. Björklund Jansson M. Two methods of extraction and GC-analysis of lipophilic
wood extractives. Report #3, Nordic Standardization Program (NSP) (2008).
27. Fernández de Simón B, García-Vallejo MC, Cadahía E, Arrabal C, and Cortijo M.
Analysis of lipophilic compounds in needles of Pinus pinea L. Ann For Sci 58,
449–454 (2001).
28. GuiJun W, YongHao L, Jian D, JiangHong K, and YunLiang Z. Pretreatment of
biomass by torrefaction. Chin Sci Bull 56(14), 1442–1448 (2011).
29. Kojiro K, Miki T, Sugimoto H, and Kanayama K. Destabilization of wood micro-
structure caused by drying. Wood Material Science & Engineering 6(1–2), 69–74
doi:10.1080/17480272.2010.551545 (2011).
30. Thuvander F, Kifetew G, and Berglund LA. Modeling of cell wall drying stresses
in wood. Wood Sci Technol 36(3), 241–254 (2002).
31. Bergman PCA and Kiel JHA. Torrefaction for biomass upgrading. In: Proceedings
of the 14th European Biomass Conference & Exhibition. Paris, France, 17–21
October 2005.
32. Bergman PCA. Combined torrefaction and pelletization: The TOP process.
Report ECN-C-05-073, ECN, Petten (2005).
33. Lam PK, Sokhansanj S, Bi T, and Lim CJ. Quantifying color of steam treated pel-
lets made from western Douglas Fir (Pseudotsuga Menziesii, L.). Trans ASABE
(under review, 2011).
34. Summa CA, de la Calle B, Brohee M, Stadler RH, and Anklam E. Impact of the
roasting degree of coffee on the in vitro radical scavenging capacity and con-
tent of acrylamide. LWT, 40, 1849-1854 (2007).
35. Shafizadeh F and McGinnis GD. Chemical composition and thermal analysis of
cottonwood. Carbohyd Res 16, 273–277 (1971).
36. Terziev N. Effect of high-temperature and microwave treatment on microstruc-
ture of softwoods. In: Proceedings of the 4th COST E15 Workshop: Methods for
Improving Drying Quality of Wood. Santiago de Compostela, Spain, 30–31 May
2002.
37. Borrega M and Kärenlampi PP. Cell wall porosity in Norway spruce wood as
affected by high temperature drying. Wood Fiber Sci 43(2), 206–214 (2011).
38. Borrega M and Kärenlampi PP. Three mechanisms affecting the mechanical prop-
erties of spruce wood dried at high temperatures. J Wood Sci 56, 87–94 (2010).
39. Weise U, Maloney T, and Paulapuro H. Quantification of water in different
states of interaction with wood pulp fiber. Cellulose 3, 189–202 (1996).
40. Park S, Venditti RA, Jameel H, and Pawlak JJ. Changes in pore size distribution
during the drying of cellulose fibers as measured by differential scanning calo-
rimetry. Carbohyd Polym 66(1), 97–103 (2006).
41. Van den Akker JA. Some theoretical considerations on the mechanical proper-
ties of fibrous structures. In: Formation and Structure of Paper. 205–241. F.
Bolam (ed.) Transactions of the Fundamental Research Symposium, Oxford,
United Kingdom, September 1961. Technical Section of the British Paper and
Board’s Makers Association, London, United Kingdom (1962).
42. Kifetew G, Thuvander F, Berglund L, and Lindberg H. The effect of drying on
wood fracture surfaces from specimens loaded in wet condition. Wood Sci
Technol 32(2), 83–94 (1998).
43. Thuvander F, Wallström L, Berglund LA, and Lindberg KAH. Effects of an
impregnation procedure for prevention of wood cell wall damage due to dry-
400 INdustrial Biotechnology october 2011
ing. Wood Sci Technol 34(6), 473–480 (2000).
44. Kauman WG. Contribution to the theory of cell collapse in wood: investigations
with eucalyptus regnans. Maderas. Ciencia y tecnología 4(1), 77–99 (2002).
45. Tiemann HD. Eucalyptus lumber. (Abstr. Rep. For. Prod. Lab., Madison,
Wisconsin) Hardwood Rec Sept. 25, Oct. 10, 1913.
46. Greenhill WL. Collapse and its removal: some recent investigations with euca-
lyptus regnans. Coun Sci Industr Res (Aust.) Pamph. 75 (1938).
47. Ellwood EL. The seasoning of rotary peeled veneer from Eucalyptus regnans
F.v.M. Aust J Appl Sci 3(1), 53–70 (1952).
48. Ellwood EL, Gottstein, JW, and Kauman WG. A laboratory study of the vapour
drying process: Part III. Vapour drying of timber in joinery and railway sleeper
sizes. CSIRO (Aust.), Div. For. Prod. Technol. Paper No. 14 (1961).
49. Demirbas A. Pyrolysis mechanisms of biomass materials. Energy Sources Part A
31(13), 1186–1193. doi: 10.1080/15567030801952268 (2009).
50. Bourgeois J and Guyonnet R. Characterization and analysis of torrefied wood.
Wood Sci Technol 22(2), 143–155 (1988).
51. Evans RJ, Milne TA, and Soltys MN. Direct mass-spectrometric studies of the
pyrolysis of carbonaceous fuels: Part III. Primary pyrolysis of lignin. J Anal Appl
Pyrol 9(3), 207–236 (1986).
52. Deng J, Wang GJ, Kuang JH, Zhang YL, and Luo YH. Pretreatment of agricultural resi-
dues for co-gasification via torrefaction. J Anal Appl Pyrol 86(2), 331–337 )2009).
53. Kiel JHA. ECN BO2-technology for biomass upgrading. BUS final meeting.
Wageningen, The Netherlands, 20 November 2007 (2007a).
54. Kiel JHA. Torrefaction for biomass upgrading into commodity fuels. In:
Proceedings of the IEA Bioenergy Task 32 Workshop on Fuel Storage, Handling
and Preparation and System Analysis for Biomass Combustion Technologies.
Berlin, Germany, 7 May 2007 (2007b).
55. White RH and Dietenberger MA. Wood products: Thermal degradation and fire.
In: The Encyclopedia of Materials: Science and Technology. Buschow KHJ, Cahn
RW, Flemings MC, Ilschner B, Kramer EJ, Mahajan S, and Veyssière P (eds.),
Elsevier Applied Science, Amsterdam, The Netherlands (2001).
56. Oliveira-Rodrigues T and Rousset PLA. Effects of torrefaction on energy proper-
ties of eucalyptus grandis wood. Cerne 15(4), 446–452 (2009).
57. Phanphanich M and Sudhagar M. Impact of torrefaction on the grindability and
fuel characteristics of forest biomass. Bioresource Technol 102(22), 1246–1253
(2011).
58. Repellin V, Govin A, Rolland M, and Guyonnet R. Energy requirement for fine
grinding of torrefied wood. Biomass Bioenerg 34(7), 923–930 (2010).
59. Mani S, Tabil LG, and Sokhansanj S. Grinding performance and physical prop-
erties of wheat and barley straws, corn stover, and switchgrass. Biomass
Bioenergy 27(4), 339–352 (2004).
60. Esteban LS and Carrasco JE. Evaluation of different strategies for pulverization
of forest biomasses. Powder Technol 166(3), 139–151 (2006).
61. Lehtikangas P. Quality properties of fuel pellets from forest biomass. Licentiate
Thesis, Report #4. Department of Forest Management and Products, University
of Agricultural Sciences. Uppsala, Sweden (1999).
62. Reed TB and Bryant B. Densified biomass: A new form of solid fuel. Solar
Energy Research Institute Report #SERI–35, US Department of Energy, Division
of Solar Technology, Golden, CO (1978).
63. Koukios EG. Progress in thermochemical, solid state refining of biofuels: From
research to commercialization. Advanced Thermochem Biomass Conversion, 2,
(1993).
64. Bridgeman TG, Darvell LI, Gudka B, Fisher EM, Jones JM, Waldron D, and
Williams A. Milling and combustion properties of torrefied biomass: A potential
co-firing fuel. Workshop on Cofiring Biomass with Coal, Session 4: Torrefaction.
Drax Power Station, United Kingdom, 25–26 January 2011.
65. Bourgeois J, Bartholin MC, and Guyonnet R. Thermal treatment of wood: analy-
sis of the obtained product. Wood Sci Technol 23, 303–310 (1989).

66. Prins MJ. Thermodynamic analysis of biomass gasification and torrefaction.
Thesis submitted to Eindhoven. Technische Universiteit Eindhoven (2005).
67. Kuang X, Tumuluru JS, Bi XT, Lim CJ, Sokhansanj S, and Melin S. Rate and peak
concentrations of off-gas emissions in stored wood pellets — sensitivities to
temperature, relative humidity, and headspace volume. Ann Occup Hyg 53(8),
789–796 (2009).
68. Tumuluru JS, Kuang X, Sokhansanj S, Lim CJ, Bi X, and Melin S. Development of lab-
oratory studies on the off-gassing of wood pellets. Can Bio Eng 52, 8.1–8.9 (2010).
69. Pastorova I, Arisz PW, and Boon JJ. Preservation of D-glucose oligosaccharides
in cellulose chars. Carbohyd Res 248, 151–165 (1993).
© mary ann liebert, inc.
research
70. Wooten JB, Crosby B, and Hajaligol MR. Evaluation of cellulose char structure moni-
tored by 13C CP MAS NMR. Fuel Chem Division Preprints 46, 191–193 (2000).
71. Sokhansanj S, Peng J, Lim J, Bi X, Wang L, Lam P, Hoi J, Melin S, Tumuluru J,
and  Wright C. Optimum torrefaction and pelletization of biomass feedstock.
TCS 2010 Symposium on Thermal and Catalytic Sciences for Biofuels and
Biobased Products. Iowa State University, Ames, Iowa, 21–23 September 2010.
72. Kleinschmidt CP. Overview of international developments on torrefaction.
Central European Biomass Conference 2011, Graz, Austria, January 26–28, 2011
[Online]. Available at: http://www.bioenergytrade.org/downloads/grazklein-
schmidtpaper2011.pdf [August 3, 2011] (2011).
• Vol. 7 No. 5 • october 2011 INdustrial Biotechnology 401